比特流协议分类模型
发布时间：2019-07-04 19:36
【摘要】：针对比特流协议分类困难的问题,提出一种比特流协议分类模型,该模型只利用比特流的物理取值和统计特性,不考虑协议中各个部分的语法、语义等信息来进行协议的分类。将比特流协议进行进制转换、数据单元切分、对数据单元进行词频统计,得到对应比特流流协议的数据单元频率统计图;使用基于贝叶斯理论设计的机器学算法对其进行学习,得到分类模型,将分类模型用于实际的协议分类。基于林肯实验室公布的数据集测试结果表明,该模型能较好地对比特流协议进行分类,正确率高,运行稳定、速度快。
[Abstract]:In order to solve the problem that it is difficult to classify bitstream protocol, a bitstream protocol classification model is proposed. The model only makes use of the physical value and statistical characteristics of bitstream, and does not consider the syntax, semantics and other information of each part of the protocol to classify the protocol. The bit flow protocol is converted, the data unit is segmented, and the word frequency statistics of the data unit are carried out, and the data unit frequency statistics diagram of the corresponding bit flow protocol is obtained. The machine algorithm based on Bayesian theory is used to learn it, and the classification model is obtained, and the classification model is applied to the actual protocol classification. The test results based on the dataset published by Lincoln Laboratory show that the model can be classified better than the special flow protocol, with high accuracy, stable operation and fast speed.
【作者单位】： 中国工程物理研究院计算机应用研究所;电子科技大学计算机科学与工程学院;
【分类号】：TP393
本文编号：2510171
[Abstract]:In order to solve the problem that it is difficult to classify bitstream protocol, a bitstream protocol classification model is proposed. The model only makes use of the physical value and statistical characteristics of bitstream, and does not consider the syntax, semantics and other information of each part of the protocol to classify the protocol. The bit flow protocol is converted, the data unit is segmented, and the word frequency statistics of the data unit are carried out, and the data unit frequency statistics diagram of the corresponding bit flow protocol is obtained. The machine algorithm based on Bayesian theory is used to learn it, and the classification model is obtained, and the classification model is applied to the actual protocol classification. The test results based on the dataset published by Lincoln Laboratory show that the model can be classified better than the special flow protocol, with high accuracy, stable operation and fast speed.
【作者单位】： 中国工程物理研究院计算机应用研究所;电子科技大学计算机科学与工程学院;
【分类号】：TP393
【相似文献】
相关会议论文 前1条
1 刘敏;张炜;;一种基于FPGA的在线更新配置方案的研究与实现[A];2008年中国高校通信类院系学术研讨会论文集（下册）[C];2009年
相关硕士学位论文 前4条
1 刘刚;比特产品定价研究[D];复旦大学;2014年
2 马良;H.264比特流解析器的设计[D];华中科技大学;2009年
3 张泱博;比特产品多维度分类研究[D];北京邮电大学;2014年
4 王宁;基于“鱼眼”降比特的视频可分级编码与传输控制研究[D];西南交通大学;2006年
，本文编号：2510171
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2510171.html
